[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <67f1741d-343c-f704-b5a8-3bc84e1c0e7e@gmail.com>
Date: Fri, 29 Jul 2022 10:40:24 +0900
From: Takahiro Kuwano <tkuw584924@...il.com>
To: Michael Walle <michael@...le.cc>,
Tudor Ambarus <tudor.ambarus@...rochip.com>,
Pratyush Yadav <p.yadav@...com>,
Miquel Raynal <miquel.raynal@...tlin.com>,
Richard Weinberger <richard@....at>,
Vignesh Raghavendra <vigneshr@...com>
Cc: linux-kernel@...r.kernel.org, linux-mtd@...ts.infradead.org
Subject: Re: [PATCH 5/6] mtd: spi-nor: add generic flash driver
On 5/13/2022 10:35 PM, Michael Walle wrote:
> Our SFDP is parsing is everything we need to support all basic
redundant "is"?
> operations of a flash device. If the flash isn't found in our in-kernel
> flash database, gracefully fall back to a driver described solely by its
> SFDP tables.
>
> It is still recommended to add the flash to the in-kernel database.
> First, we get a proper partname and secondly, for all features not
> described by the SFDP like OTP we need the entry anyway.
>
> Signed-off-by: Michael Walle <michael@...le.cc>
> ---
> drivers/mtd/spi-nor/core.c | 13 +++++++++++++
> drivers/mtd/spi-nor/core.h | 1 +
> drivers/mtd/spi-nor/sfdp.c | 27 +++++++++++++++++++++++++++
> 3 files changed, 41 insertions(+)
>
> diff --git a/drivers/mtd/spi-nor/core.c b/drivers/mtd/spi-nor/core.c
> index 65cd8e668579..ee193a61310a 100644
> --- a/drivers/mtd/spi-nor/core.c
> +++ b/drivers/mtd/spi-nor/core.c
> @@ -1632,6 +1632,11 @@ static const struct spi_nor_manufacturer *manufacturers[] = {
> &spi_nor_xmc,
> };
>
> +static const struct flash_info spi_nor_generic_flash = {
> + .name = "spi-nor-generic",
> + .parse_sfdp = true,
> +};
> +
> static const struct flash_info *spi_nor_match_id(struct spi_nor *nor,
> const u8 *id)
> {
> @@ -1670,6 +1675,14 @@ static const struct flash_info *spi_nor_detect(struct spi_nor *nor)
> return ERR_PTR(-ENOMEM);
>
> info = spi_nor_match_id(nor, id);
> +
> + /* Fallback to a generic flash described only by its SFDP data. */
> + if (!info) {
> + ret = spi_nor_check_sfdp_signature(nor);
> + if (!ret)
> + info = &spi_nor_generic_flash;
> + }
> +
> if (!info) {
> dev_err(nor->dev, "unrecognized JEDEC id bytes: %*ph\n",
> SPI_NOR_MAX_ID_LEN, id);
> diff --git a/drivers/mtd/spi-nor/core.h b/drivers/mtd/spi-nor/core.h
> index 3a19b8092ab8..aa9f218245a5 100644
> --- a/drivers/mtd/spi-nor/core.h
> +++ b/drivers/mtd/spi-nor/core.h
> @@ -694,6 +694,7 @@ int spi_nor_controller_ops_read_reg(struct spi_nor *nor, u8 opcode,
> int spi_nor_controller_ops_write_reg(struct spi_nor *nor, u8 opcode,
> const u8 *buf, size_t len);
>
> +int spi_nor_check_sfdp_signature(struct spi_nor *nor);
> int spi_nor_parse_sfdp(struct spi_nor *nor);
>
> static inline struct spi_nor *mtd_to_spi_nor(struct mtd_info *mtd)
> diff --git a/drivers/mtd/spi-nor/sfdp.c b/drivers/mtd/spi-nor/sfdp.c
> index a5211543d30d..9bdb3d5dc7e8 100644
> --- a/drivers/mtd/spi-nor/sfdp.c
> +++ b/drivers/mtd/spi-nor/sfdp.c
> @@ -1247,6 +1247,33 @@ static void spi_nor_post_sfdp_fixups(struct spi_nor *nor)
> nor->info->fixups->post_sfdp(nor);
> }
>
> +/**
> + * spi_nor_check_sfdp_header() - check for a valid SFDP header
> + * @nor: pointer to a 'struct spi_nor'
> + *
> + * Used to detect if the flash supports the RDSFDP command as well as the
> + * presence of a valid SFDP table.
> + *
> + * Return: 0 on success, -errno otherwise.
> + */
> +int spi_nor_check_sfdp_signature(struct spi_nor *nor)
> +{
> + u32 signature;
> + int err;
> +
> + /* Get the SFDP header. */
> + err = spi_nor_read_sfdp_dma_unsafe(nor, 0, sizeof(signature),
> + &signature);
> + if (err < 0)
> + return err;
> +
> + /* Check the SFDP signature. */
> + if (le32_to_cpu(signature) != SFDP_SIGNATURE)
> + return -EINVAL;
> +
> + return 0;
> +}
> +
Nice to use this function from spi_nor_parse_sfdp() as well, but I found it's
not straightforward...
Reviewed-by: Takahiro Kuwano <Takahiro.Kuwano@...ineon.com>
Thanks,
Takahiro
Powered by blists - more mailing lists